SWFPS: PROPOSITION OF A DATA AND PROCESSES PROVENANCE SYSTEM IN THE DOMAIN OF SCIENTIFIC WORKFLOWS

Abstract

This paper describes SWfPS, an architecture that aims to interact with Scientific Workflow Management Systems in order to capture and manipulate provenance metadata. For this purpose, SWfPS adopts an approach based on an abstract model for representing the lineage. This model, called Open Provenance Model, allows SWfPS to set up a homogeneous and interoperable infrastructure for handling provenance metadata. As a result, SWfPS is able to provide a framework for query metadata provenance generated in an e-Science scenario. Moreover, the architecture uses semantic web technology in order to process provenance queries. In this context, using ontologies and inference engines, SWfPS can make inferences about lineage and, based on those inferences, obtain important results from the extraction of information from the managed data, beyond what is explicitly registered.